INDEX

Explanations

names of organizations or positions related to government, corporate, or nonprofit entities

oai_token-act-pair · gpt-3.5-turbo

terms related to various professional roles and organizations

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.hook_resid_pre

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

torch.float32

Hook Point

blocks.10.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 deval

-0.64

DragonMagazine

-0.63

?),

-0.62

 stret

-0.62

 hath

-0.61

 snippets

-0.60

 reverted

-0.57

 nude

-0.57

 doesnt

-0.57

theless

-0.57

POSITIVE LOGITS

_.

1.01

0.96

.ï¿½

0.94

>.

0.93

.''

0.90

.</

0.90

.''.

0.89

}.

0.82

.(

0.81

.*

0.81

Activations Density 0.338%

names of organizations or positions related to government, corporate, or nonprofit entities

terms related to various professional roles and organizations

No Comments

No Known Activations